Duty Cycle stuck on 90% - 89 300SEL

[Durbo20vT]

I've been battling with finding the cause of this fault & have tried some tests found from searching BW without success. The car is an 89 M103, non-California car, so with my DMM plugged in ports 2&3, I should see 70%... but I do not.

My car reads 90.5% at all times from key on engine off, all the way thru running to 80* C. Flipping the poles in the diag port, i get 9.5% from KOEO thru running to full temp. The only time I get a fluctuation is upon throttle enrichment, where it moves to 60% then immediately back down to 90.5% upon liftoff. The car will start rough and die first attempt in the morning, but then starts to a normal idle of 750 during warm up phase & move to 1100-1200 rpm idle after reaching 80*C and stays there.

I've eliminated vac leaks & changed a ton of parts (list below). Hoping for a change, I unplugged the o2 sensor, deflected the air plate, tested voltage to IAC(11.7v)/potentiometer/EHA & all appeared to be well but did not see any change other than the CEL lighting up. No codes are stored when pulled with the code reader, just the 90.5% duty cycle (or 9.6%)...

The car is in what seems to be "limp home mode" & I'm looking to get out of it & adjust my duty cycle down to the 50% it should be at! I'm at the point where I've exhausted throwing parts at the problem, so what could I be overlooking which would correct this fault?


List of parts changed without success:

Brand new KAE OVPs (latest one 2 weeks old)
New Bosch 13925 o2 Sensor (last year)
New Potentiometer
New Fuel Pump & Filter
New Bosch Injectors & seals
Replaced IAC hoses
New Airflow meter boot
New throttle microswitch
Swapped throttle body+sensor with known working unit (new gasket)
Swapped EHA with a different unit
New Bosch Voltage Regulator
New Belt/Tensioner & associated parts
New Radiator (no impact expected)
Bosch Cap, Suppressor and rotor (1 year old)
Denso spark plugs (1 year old)
New 2-pin temp sensor (1 year old)
Swapped cracked hose from valve cover to breather tube on cyl head
New Bosch Coil
Replaced Fuel Pressure Regulator

[Stretch]

Go through all of the tedious tests described in chapter 07.3-0121

Below is a link for a W124 300E - but there's a similar if not the same chapter in the W126 FSM that is also available on www.startekinfo.com

http://www.startekinfo.com/StarTek/outside/12264/Program/Engine/103/07.3-0121.pdf

For hints and tips to get that FSM to work see link in my sig...

[TnBob]

I see you have been having 'fun' too !

[Durbo20vT]

Tons! Lol

Thanks for the link, I began a few tests before it got too dark. EHA wiring tested within spec at 19.5 on my DMM (under mA).

Idle control valve sees voltage at the connector, but unplugging/plugging the icv while car is running provides no change... Stuck at high idle.

I did find this fuse frail and cracked by the fuse box... Unsure what it controls but here's where I can start replacing...

[Stretch]

I don't know what that fuse is for - possibly something big like AC?

From what I've read so far a disconnected air idle control valve should provide a high idle - connect it and it should work and the idle should drop (unless the car is in the warm up phase). Skip to the section for checking that valve - if you briefly apply 12V to the correct pins (check!) it should click...

[Elektri]

On my 89 W126 KE-Jetronic, the duty cycle is supposed to be about 50% when warm. And I use a multimeter which has a duty cycle setting. I then need to reverse the leads as my multimeters reads % ON for Duty Cycle - Mercedes uses % OFF.

With that said, once I get an accurate duty cycle reading, the engine computer will be "stuck" at a certain duty cycle reading if there is a malfunction. This is similar to the duty cycle percentages in the following document. I don't know if this applies to your specific car? Here is that document... (See page 16.)

http://www.berlinasportivo.com/Technical/lancia/Thema832/Testing%20the%20KE%20Injection%20System.pdf

P.S. I just searched google for any old document which had the % readouts. I don't know which specific vehicle that document applies to, but may want to find a similar document in the factory service manual for your specific vehicle.

[Elektri]

Also that fuse probably goes to the A/C / Heating blower motor inside the car.

[92 300D]

Correct the fuse does go to blower motor. I believe it's mentioned on fuse card.

Sent from my SM-G900V using Tapatalk

[timm9]

I'm assuming that it is running rich?

Not sure if it is the same problem but I spent weeks trying to find a similar issue with my 82 380SL that turned out to be cracks in the air chamber which could not be seen and my duty cycle was at 90%. I only found it after doing a smoke test. The cracks were allowing additional air into the system which told the system to add more fuel. Unfortunately, I had replaced many unnecessary parts in the process. if you think there is a chance it is the same issue there are posts on this site with photos.

[Durbo20vT]

Originally Posted by timm9

I'm assuming that it is running rich?

Not sure if it is the same problem but I spent weeks trying to find a similar issue with my 82 380SL that turned out to be cracks in the air chamber which could not be seen and my duty cycle was at 90%. I only found it after doing a smoke test. The cracks were allowing additional air into the system which told the system to add more fuel. Unfortunately, I had replaced many unnecessary parts in the process. if you think there is a chance it is the same issue there are posts on this site with photos.

Did yours fluctuate at all, or was it permanently on 90% ? Mine is permanently on 90 (or 9% if poles are flipped), regardless if car is on or key on engine off.

I am trying some more tests this evening & tracked down a newer idle control valve. The ICV which I have clicks, but testing my current unit seemed a bit off.

[TnBob]

Elektri, the pdf file link you posted has a pretty major w124 error. It shows the O2 sensor at the right rear of the block. THAT is the location of the coolant temperature sensor.

My O2 sensor is screwed into the exhaust and the cts screws into the block

[TnBob]

These M103's can be a nitemare to diagnose primarily due to all the different parts the factory threw at the motor.

Some of the coolant temperature sensors have an ohm reading of 3k+ while others have a sub 50 ohm reading, both values are on a cold motor as an example.

Some M103's require removing the valve cover to access the coolant temperature sensor, thankfully mine doesnt... the 22mm socket dropped right on.

Some ECU's or whatever does the adjustment allow for the Idle Air Valve in front of the fuel distributor to raise the idle when disconnected, my M103 on two different units dropped the idle speed by at least 500rpm at startup from cold.

Some EHA's have a 2mm set screw adjustment, others have a flat head screw.

The upshot is the controlling or adjusting the startup idle speed on this motor is greatly lacking in fun.

[MBeige]

I've had a similar issue with my former '89 190E 2.6.

O2 sensor was causing chaos with the duty cycle, IIRC. Basically it rendered the adjustment non functional because there was no longer a loop/feedback from the sensor.



However, the EHA was also causing high idle at operating temperature. Bought a new Bosch unit and fixed the problem. The car is gone but I still have the EHA.

Even with a non-leaking EHA, if it was not good it would wreak havoc on the idle and throttle response.

[Durbo20vT]

Originally Posted by MBeige

I've had a similar issue with my former '89 190E 2.6.

O2 sensor was causing chaos with the duty cycle, IIRC. Basically it rendered the adjustment non functional because there was no longer a loop/feedback from the sensor.



However, the EHA was also causing high idle at operating temperature. Bought a new Bosch unit and fixed the problem. The car is gone but I still have the EHA.

Even with a non-leaking EHA, if it was not good it would wreak havoc on the idle and throttle response.

thanks for the input! After some tests to the temp sensor/wiring, FPR, and ECU, I'm leaning toward replacing the actual EHA as you did above. I have a constant 90% duty cycle with absolutely no fluctuation, high idle after reaching operating temp and lacking in power toward the upper rpm range. I will try and track down a known working EHA to verify this is the root cause prior to taking the plunge and buying new.


Temp Sensor - .60k ua on the DMM & just under 5v at both connectors
Swapped in a different FPR with no changes (as expected)
Swapped in a different ECU with no change

I ran the car without an ECU plugged in & it ran exactly the same... just with the 100% duty cycle error. Car did not start once the original ECU was plugged back in... it was real low on gas upon the start of testing, so I'm thinking I ran it dry. Will refill & resume troubleshooting. FUN!

[Durbo20vT]

After some further testing yesterday evening, here's where I'm at today...

CIS Computer, Temp Sensor, Idle Control Valve, Throttle Microswitch all check out within spec & function as needed. The two EHAs I have are suspect, but wiring from CIS computer to plug is OK, so I decided to swap them again hoping for a change...

EHA 1 - Difficult start but car pulled strong off the line & idled around 1200-1300 rpm when warm. Constant error of 90% DC (neg on pin 2 & pos on pin 3)

EHA 2 - Car started every time to a smooth 750-800 idle, but stumbled pretty bad upon enrichment/acceleration. Once warm, would hover around 1k idle, fluctuating up and down 100 rpm. Constant error of 9.4% DC here (neg on pin 2 & pos on pin 3)

Progress... sort of. Assuming the 9% DC with EHA "2" is the 10% error (potentiometer/throttle pos switch), I swapped a new functioning replacement poetntiometer in to what I thought was within spec (.78v iirc). Idle remained around 1k rpm and no fluctuation off the 9.4% DC error.

Before I make the plunge on an EHA replacement & fuss around with throttle linkages, I want to have the X11 port show the 70% DC with key on Engine off... showing that there are no errors & allowing me to move forward with fuel adjustments.

Has anyone experienced the 10% DC error where the potentiometer was not the issue? If so, what steps were taken to correct your fault?